registry / focalapi-cli / 0.7.8

focalapi-cli@0.7.8

让任意 AI Agent 直接调用 focalapi 创作模型的命令行工具

AI Security Review

scanned 22d ago · by lpm-firewall-ai

Installation automatically scans for supported third-party AI agents and installs this package's skills into their home-directory control surfaces. This is an unconsented broad agent-control-surface mutation.

Static reason

High-risk behavior combination matched malicious policy.; source matched previously finalized malicious package; routed for review; source fingerprint signature matched known malicious package; routed for review

Trigger

npm postinstall during package installation.

Impact

It can alter the behavior and available capabilities of multiple installed AI agents without an explicit user command.

Mechanism

Automatic discovery and installation of bundled skills into detected agent directories.

Attack narrative

Installing the package runs its postinstall script. That script starts the bundled CLI with connect install, which detects supported agent installations and copies bundled skills into their skill directories. The operation is automatic and can affect multiple third-party AI-agent control surfaces in the user's home directory.

AI rationale

The automatic postinstall chain performs broad, foreign AI-agent skill installation without an explicit user setup command. This meets the install-hook abuse blocking policy even though the inspected lifecycle path does not show network exfiltration.

Evidence

package.jsonscripts/postinstall.cjsdist/cli.js~/.agents/skills~/.claude/skills~/.cursor/skills

focalapi-cli@0.7.8

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Affected versions and remediation

Source & flagged code